|Publication number||US6658015 B1|
|Application number||US 09/321,623|
|Publication date||Dec 2, 2003|
|Filing date||May 28, 1999|
|Priority date||May 28, 1999|
|Publication number||09321623, 321623, US 6658015 B1, US 6658015B1, US-B1-6658015, US6658015 B1, US6658015B1|
|Inventors||Shashank Merchant, Chandan Egbert, John M. Chiang|
|Original Assignee||Advanced Micro Devices, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (7), Referenced by (30), Classifications (10), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present invention relates to network communications and more particularly, to generating data forwarding decisions in a network switch.
In computer networks, a plurality of network stations are interconnected via a communications medium. For example, Ethernet is a commonly used local area network scheme in which multiple stations are connected to a single shared serial data path. These stations often communicate with a switch located between the shared data path and the stations connected to that path. Typically, the switch controls the communication of data packets on the network.
The network switch includes switching logic for receiving and forwarding frames to the appropriate destinations. One arrangement for generating a frame forwarding decision uses a direct addressing scheme, where the network switch includes a fixed address table storing switching logic for the destination addresses.
For example, a frame may be received by the network switch with header information indicating the source address and destination address of the frame. The switching logic accesses the fixed address table using the source address and destination address as lookups to find the appropriate frame forwarding information. The switch then uses this information and sends the frame to the appropriate port(s).
When all of the stations connected to the network are simultaneously operating, packet traffic on the shared serial path can be heavy with little time between packets. Additionally, due to increased network throughput requirements, increasing the data throughput is crucial to the overall operation of the switch.
One drawback with typical prior art systems employing address tables is that the decision making logic processes data frames one frame at a time. That is, when multiple frames are received by the switch, the decision making logic processes the header information from the first frame and generates the frame forwarding information. The decision making logic then starts the process over again for the second frame. Such a processing arrangement often results in low data throughput. Additionally, prior art arrangements employing address tables typically include a single decision making device that performs all of the processing tasks necessary to generate a frame forwarding decision. A drawback with utilizing a single decision making device is the difficulty in implementing changes in any one part of the decision making process without affecting other parts of the decision making process. Another drawback with typical prior art systems is that the decision making logic is included within a logic device that performs other functions on the switch. Such an arrangement makes changes to the decision making logic difficult to implement without affecting other logic functions on the switch.
There exists a need for a switching device that includes a decision making engine designed to support networks requiring a high data throughput.
There is also a need for a switching device that employs a modular decision making engine that facilitates changes to the decision making logic.
These and other needs are met by the present invention, where a multiport switch includes a decision making engine used to make frame forwarding decisions. The decision making engine is designed in pipelined, modular fashion so that multiple frames may be processed simultaneously.
According to one aspect of the invention, a multiport switch is configured to control the communication of data frames between stations. The switch includes a programmable address table for storing address information and data forwarding information. The switch also includes a decision making engine configured to search the programmable address table and generate data forwarding information for a data frame. The decision making engine comprises a plurality of logic engines where each of the logic engines is configured to process a data frame simultaneously with each other logic engine.
Another aspect of the present invention provides a method for generating data forwarding information in a multiport switch that controls communication of data frames between stations. The method includes receiving a data frame and processing the data frame in a pipelined manner through a decision making engine. The decision making engine comprises a plurality of logic engines with each of the logic engines configured to process a data frame simultaneously with each other logic engine. The method also includes generating the data forwarding information.
Other advantages and features of the present invention will become readily apparent to those skilled in this art from the following detailed description. The embodiments shown and described provide illustration of the best mode contemplated for carrying out the invention. The invention is capable of modifications in various obvious respects, all without departing from the invention. Accordingly, the drawings are to be regarded as illustrative in nature, and not as restrictive.
FIG. 1 is a block diagram of a packet switched network including a multiple port switch according to an embodiment of the present invention.
FIG. 2 is a block diagram of the multiple port switch of FIG. 1.
FIG. 3 is a detailed block diagram illustrating the switching subsystem of FIG. 2.
FIG. 4 is a block diagram of a system including the internal rules checker of FIG. 2 in accordance with an embodiment of the present invention.
FIG. 5 illustrates the composition of the IRC address table of FIG. 4.
FIG. 6 illustrates the format of an IRC address table entry of the IRC address table of FIG. 5.
FIG. 7 illustrates linked list chains for identifying table entries relative to a selected bin. FIG. 8 illustrates a hash function circuit used with the internal rules checker of FIG. 2.
FIG. 9 illustrates the composition of the forwarding descriptor in accordance with an embodiment of the present invention.
The present invention will be described with the example of a switch in a packet switched network, such as an Ethernet (IEEE 802.3) network. It will become apparent, however, that the present invention is also applicable to other packet switched systems, as described in detail below, as well as to other types of systems in general.
FIG. 1 is a block diagram of an exemplary system in which the present invention may be advantageously employed. The exemplary system 10 is a packet switched network, such as an Ethernet (IEEE 802.3) network. The packet switched network includes integrated multiport switches (IMS) 12 that enable communication of data packets between network stations. The network may include network stations having different configurations, for example twelve (12) 10 megabit per second (Mb/s) or 100 Mb/s network stations 14 (hereinafter 10/100 Mb/s) that send and receive data at a network data rate of 10 Mb/s or 100 Mb/s, and a 1000 Mb/s (i.e., 1 Gb/s) network node 22 that sends and receives data packets at a network speed of 1 Gb/s. The gigabit node 22 may be a server, or a gateway to a high-speed backbone network. Hence, the multiport switches 12 selectively forward data packets received from the network nodes 14 or 22 to the appropriate destination based upon Ethernet protocol.
Each multiport switch 12 includes a media access control (MAC) module 20 that transmits and receives data packets to and from 10/100 Mb/s physical layer (PHY) transceivers 16 via respective shared media independent interfaces (MII) 18 according to IEEE 802.3u protocol. Each multiport switch 12 also includes a gigabit MAC 24 for sending and receiving data packets to and from a gigabit PHY 26 for transmission to the gigabit node 22 via a high speed network medium 28.
Each 10/100 Mb/s network station 14 sends and receives data packets to and from the corresponding multiport switch 12 via a media 17 and according to either half-duplex or full duplex Ethernet protocol. The Ethernet protocol ISO/IEC 8802-3 (ANSI/IEEE Std. 802.3, 1993 Ed.) defines a half-duplex media access mechanism that permits all stations 14 to access the network channel with equality. Traffic in a half-duplex environment is not distinguished or prioritized over the medium 17. Rather, each half-duplex station 14 includes an Ethernet interface card that uses carrier-sense multiple access with collision detection (CSMA/CD) to listen for traffic on the media. The absence of network traffic is detected by sensing deassertion of a receive carrier on the media. Any station 14 having data to send will attempt to access the channel by waiting a predetermined time, known as the interpacket gap interval (IPG), after deassertion of the receive carrier on the media. If a plurality of stations 14 have data to send on the network, each of the stations will attempt to transmit in response to the sensed deassertion of the receive carrier on the media and after the IPG interval, possibly resulting in a collision. Hence, the transmitting station will monitor the media to determine if there has been a collision due to another station sending data at the same time. If a collision is detected, both stations stop, wait a random amount of time, and retry transmission.
The 10/100 Mb/s network stations 14 that operate in full duplex mode send and receive data packets according to the Ethernet standard IEEE 802.3u. The full-duplex environment provides a two-way, point-to-point communication link enabling simultaneous transmission and reception of data packets between each link partner, i.e., the 10/100 Mb/s network station 14 and the corresponding multiport switch 12.
Each multiport switch 12 is coupled to 10/100 physical layer (PHY) transceivers 16 configured for sending and receiving data packets to and from the corresponding multiport switch 12 across a corresponding shared media independent interface (MII) 18. In particular, each 10/100 PHY transceiver 16 is configured for sending and receiving data packets between the multiport switch 12 and up to four (4) network stations 14 via the shared MII 18. A magnetic transformer 19 provides AC coupling between the PHY transceiver 16 and the corresponding network medium 17. Hence, the shared MII 18 operates at a data rate sufficient to enable simultaneous transmission and reception of data packets by each of the network stations 14 to the corresponding PHY transceiver 16.
Each multiport switch 12 also includes an expansion port 30 for transferring data between other switches according to a prescribed protocol. For example, each expansion port 30 can be implemented as a second gigabit MAC port similar to port 24, thereby enabling multiple multiport switches 12 to be cascaded together as a separate backbone network.
FIG. 2 is a block diagram of the multiport switch 12. The multiport switch 12 contains a decision making engine 40 that performs frame forwarding decisions, a switching subsystem 42 for transferring frame data according to the frame forwarding decisions, an external memory interface 44, management information base (MIB) counters 48 a and 48 b (collectively 48), and MAC (media access control) protocol interfaces 20 and 24 to support the routing of data packets between the Ethernet (IEEE 802.3) ports serving the network stations 14 and the gigabit node 22. The MIB counters 48 provide statistical network information in the form of management information base (MIB) objects, to an external management entity controlled by a host CPU 32, described below.
The external memory interface 44 enables external storage of packet data in an external memory 36 such as, for example, a synchronous static random access memory (SSRAM), in order to minimize the chip size of the multiport switch 12. In particular, the multiport switch 12 uses the external memory 36 for storage of received frame data, memory structures, and MIB counter information. The external memory 36 is preferably either a Joint Electron Device Engineering Council (JEDEC) pipelined burst or Zero Bus Tumaround™ (ZBT)-SSRAM having a 64-bit wide data path and a 17-bit wide address path. The external memory 36 is addressable as upper and lower banks of 128K in 64-bit words. The size of the external memory 36 is preferably at least 1 Mbytes, with data transfers possible on every clock cycle through pipelining. Additionally the external memory interface clock operates at clock frequencies of at least 66 MHz, and, preferably, 100 MHz and above.
The multiport switch 12 also includes a processing interface 50 that enables an external management entity such as a host CPU 32 to control overall operations of the multiport switch 12. In particular, the processing interface 50 decodes CPU accesses within a prescribed register access space, and reads and writes configuration and status values to and from configuration and status registers 52.
The internal decision making engine 40, referred to as an internal rules checker (IRC), makes frame forwarding decisions for data packets received from one source to at least one destination station.
The multiport switch 12 also includes an LED interface 54 that clocks out the status of conditions per port and drives an external LED logic. The external LED logic drives LED display elements that are human readable.
The switching subsystem 42, configured for implementing the frame forwarding decisions of the IRC 40, includes a port vector first in first out (FIFO) buffer 56, a plurality of output queues 58, a multicopy queue 60, a multicopy cache 62, a free buffer queue 64, and a reclaim queue 66.
The MAC unit 20 includes modules for each port, each module including a MAC receive portion, a receive FIFO buffer, a transmit FIFO buffer, and a MAC transmit portion. Data packets from a network station 14 are received by the corresponding MAC port and stored in the corresponding receive FIFO. The MAC unit 20 obtains a free buffer location (i.e., a frame pointer) from the free buffer queue 64, and outputs the received data packet from the corresponding receive FIFO to the external memory interface 44 for storage in the external memory 36 at the location specified by the frame pointer.
The IRC 40 monitors (i.e., “snoops”) the data bus to determine the frame pointer value and the header information of the received packet (including source, destination, and VLAN address information). The IRC 40 uses the header information to determine which MAC ports will output the data frame stored at the location specified by the frame pointer. The decision making engine (i.e., the IRC 40) may thus determine that a given data frame should be output by either a single port, multiple ports, or all ports (i.e., broadcast). For example, each data frame includes a header having source and destination address, where the decision making engine 40 may identify the appropriate output MAC port based upon the destination address. Alternatively, the destination address may correspond to a virtual address that the appropriate decision making engine identifies as corresponding to a plurality of network stations. In addition, the frame may include a VLAN tag header that identifies the frame as information destined to one or more members of a prescribed group of stations. The IRC 40 may also determine that the received data packet should be transferred to another multiport switch 12 via the expansion port 30. Hence, the internal rules checker 40 will decide whether a frame temporarily stored in the external memory 36 should be output to a single MAC port or multiple MAC ports.
The internal rules checker 40 outputs a forwarding decision to the switch subsystem 42 in the form of a forwarding descriptor. The forwarding descriptor includes a priority class identifying whether the frame is high priority or low priority, a port vector identifying each MAC port that should receive the data frame, Rx port number, an untagged set field, VLAN information, opcode, and frame pointer. The format of the forwarding descriptor will discussed further with respect to FIG. 9. The port vector identifies the MAC ports to receive the data frame for transmission (e.g., 10/100 MAC ports 1-12, Gigabit MAC port, and/or Expansion port). The port vector FIFO 56 decodes the forwarding descriptor including the port vector, and supplies the frame pointer to the appropriate output queues 58 that correspond to the output MAC ports to receive the data frame transmission. In other words, the port vector FIFO 56 supplies the frame pointer on a per-port basis. The output queues 58 fetch the data frame identified in the port vector from the external memory 36 via the external memory interface 44, and supply the retrieved data frame to the appropriate transmit FIFO of the identified ports. If a data frame is to be supplied to a management agent, the frame pointer is also supplied to a management queue 68, which can be processed by the host CPU 32 via the CPU interface 50.
The multicopy queue 60 and the multicopy cache 62 keep track of the number of copies of the data frame that are fetched from the respective output queues 58, ensuring that the data frame is not overwritten in the external memory 36 until the appropriate number of copies of the data frame have been output from the external memory 36. Once the number of copies output corresponds to the number of ports specified in the port vector FIFO 56, the frame pointer is forwarded to the reclaim queue 66. The reclaim queue 66 stores frame pointers that can be reclaimed by the free buffer queue 64 as free pointers. After being returned to the free buffer queue 64, the frame pointer is available for reuse by the MAC unit 20 or the gigabit MAC unit 24.
FIG. 3 depicts the switch subsystem 42 of FIG. 2 in more detail according to an exemplary embodiment of the present invention. Other elements of the multiport switch 12 of FIG. 2 are reproduced in FIG. 3 to illustrate the connections of the switch subsystem 42 to these other elements.
As shown in FIG. 3, the MAC module 20 includes a receive portion 20 a and a transmit portion 24 b. The receive portion 20 a and the transmit portion 24 b each include 12 MAC modules (only two of each shown and referenced by numerals 70 a, 70 b, 70 c, and 70 d) configured for performing the corresponding receive or transmit function according to IEEE 802.3 protocol. The MAC modules 70 c and 70 d perform the transmit MAC operations for the 10/100 Mb/s switch ports complementary to modules 70 a and 70 b, respectively.
The gigabit MAC port 24 also includes a receive portion 24 a and a transmit portion 24 b, while the expansion port 30 similarly includes a receive portion 30 a and a transmit portion 30 b. The gigabit MAC port 24 and the expansion port 30 also have receive MAC modules 72 a and 72 b optimized for the respective ports. The transmit portions 24 b and 30 b of the gigabit MAC port 24 and the expansion port 30 a also have transmit MAC modules 72 c and 72 d, respectively. The MAC modules are configured for full-duplex operation on the corresponding port, and the gigabit MAC modules 72 a and 72 c are configured in accordance with the Gigabit Proposed Standard IEEE Draft P802.3z.
Each of the receive MAC modules 70 a, 70 b, 72 a, and 72 b include queuing logic 74 for transfer of received data from the corresponding internal receive FIFO to the external memory 36 and the rules checker 40. Each of the transmit MAC modules 70 c, 70 d, 72 c, and 72 d includes a dequeuing logic 76 for transferring data from the external memory 36 to the corresponding internal transmit FIFO, and a queuing logic 74 for fetching frame pointers from the free buffer queue 64. The queuing logic 74 uses the fetched frame pointers to store receive data to the external memory 36 via the external memory interface controller 44. The frame buffer pointer specifies the location in the external memory 36 where the received data frame will be stored by the receive FIFO.
The external memory interface 44 includes a scheduler 80 for controlling memory access by the queuing logic 74 or dequeuing logic 76 of any switch port to the external memory 36, and an SSRAM interface 78 for performing the read and write operations with the external memory 36. In particular, the multiport switch 12 is configured to operate as a non-blocking switch, where network data is received and output from the switch ports at the respective wire rates of 10, 100, or 1000 Mb/s. Hence, the scheduler 80 controls the access by different ports to optimize usage of the bandwidth of the external memory 36.
Each receive MAC stores a portion of a frame in an internal FIFO upon reception from the corresponding switch port; the size of the FIFO is sufficient to store the frame data that arrives between scheduler time slots. The corresponding queuing logic 74 obtains a frame pointer and sends a write request to the external memory interface 44. The scheduler 80 schedules the write request with other write requests from the queuing logic 74 or any read requests from the dequeuing logic 76, and generates a grant for the requesting queuing logic 74 (or the dequeuing logic 76) to initiate a transfer at the scheduled event (i.e., slot). Sixty-four bits of frame data is then transferred over a write data bus 69 a from the receive FIFO to the external memory 36 in a direct memory access (DMA) transaction during the assigned slot based on the retrieved frame pointer. The frame data is stored in the location pointed to by the free buffer pointer obtained from the free buffer pool 64, although a number of other buffers may be used to store data frames, as will be described.
The rules checker 40 also receives the frame pointer and the header information (including source address, destination address, VLAN tag information, etc.) by monitoring (i.e., snooping) the DMA write transfer on the write data bus 69 a. The rules checker 40 uses the header information to make the forwarding decision and generate a forwarding instruction in the form of a forwarding descriptor that includes a port vector. The port vector has a bit set for each output port to which the frame should be forwarded. If the received frame is a unicopy frame, only one bit is set in the port vector generated by the rules checker 40. The single bit that is set in the port vector corresponds to a particular one of the ports.
The rules checker 40 outputs the forwarding descriptor including the port vector and the frame pointer into the port vector FIFO 56. The port vector is examined by the port vector FIFO 56 to determine which particular output queue should receive the associated frame pointer. The port vector FIFO 56 places the frame pointer into the top of the appropriate queue 58 and/or 68. This queues the transmission of the frame.
As shown in FIG. 3, each of the transmit MAC units 70 c, 70 d, 72 d, and 72 c has an associated output queue 58 a, 58 b, 58 c, and 58 d, respectively. In preferred embodiments, each of the output queues 58 has a high priority queue for high priority frame pointers, and a low priority queue for low priority frame pointers. The high priority frame pointers are used for data frames that require a guaranteed access latency, e.g., frames for multimedia applications or management MAC frames. The frame pointers stored in the FIFO-type output queues 58 are processed by the dequeuing logic 76 for the respective transmit MAC units. At some point in time, the frame pointer reaches the bottom of an output queue 58, for example, output queue 58 d for the gigabit transmit MAC 72 c. The dequeuing logic 76 for the transmit gigabit port 24 b takes the frame pointer from the corresponding gigabit port output queue 58 d, and issues a request to the scheduler 80 to read the frame data from the external memory 36 at the memory location specified by the frame pointer. The scheduler 80 schedules the request, and issues a grant for the dequeuirig logic 76 of the transmit gigabit port 24 b to initiate a DMA read. In response to the grant, the dequeuing logic 76 reads the frame data (along the read bus 69 b) in a DMA transaction from the location in external memory 36 pointed to by the frame pointer, and stores the frame data in the internal transmit FIFO for transmission by the transmit gigabit MAC 72 c. If the frame pointer specifies a unicopy transmission, the frame pointer is returned to the free buffer queue 64 following writing the frame data into the transmit FIFO.
A multicopy transmission is similar to the unicopy transmission, except that the port vector has multiple bits set, designating the multiple ports from which the data frame will be transmitted. The frame pointer is placed into each of the appropriate output queues 58 and transmitted by the appropriate transmit MAC units 20 b, 24 b, and/or 30 b.
The free buffer pool 64, the multicopy queue 60, the reclaim queue 66, and the multicopy ache 62 are used to manage use of frame pointers and re-use of frame pointers once the data frame has been transmitted to its designated output port(s). In particular, the dequeuing logic 76 passes frame pointers for unicopy frames to the free buffer queue 64 after the buffer contents have been copied to the appropriate transmit FIFO.
For multicopy frames, the port vector FIFO 56 supplies multiple copies of the same frame pointer to more than one output queue 58, each frame pointer having a unicopy bit set to zero. The port vector FIFO 56 also copies the frame pointer and the copy count to the multicopy queue 60. The multicopy queue 60 writes the copy count to the multicopy cache 62. The multicopy cache 62 is a random access memory having a single copy count for each buffer in external memory 36 (i.e., each frame pointer).
Once the dequeuing logic 76 retrieves the frame data for a particular output port based on a fetched frame pointer and stores the frame data in the transmit FIFO, the dequeuing logic 76 checks if the unicopy bit is set to 1. If the unicopy bit is set to 1, the frame pointer is returned to the free buffer queue 64. If the unicopy bit is set to zero indicating a multicopy frame pointer, the dequeuing logic 76 writes the frame pointer with a copy count of minus one (−1) to the multicopy queue 60. The multicopy queue 60 adds the copy count to the entry stored in the multicopy cache 62.
When the copy count in multicopy cache 62 for the frame pointer reaches zero, the frame pointer is passed to the reclaim queue 66. Since a plurality of frame pointers may be used to store a single data frame in multiple buffer memory locations, the frame pointers are referenced to each other to form a linked-list (i.e., chain) of frame pointers to identify the stored data frame in its entirety. The reclaim queue 66 traverses the chain of buffer locations identified by the frame pointers, and passes the frame pointers to the free buffer queue 64.
The foregoing description of the switch architecture provides an overview of the switch operations in a packet switched network. A more detailed description of the features of the present invention as embodied in the multiport switch 12 will now be provided.
The present invention is directed to the internal rules checker 40 (IRC) and the use of the IRC 40 to provide high data throughput. As described previously, the switch subsystem 42 provides the switching logic for receiving and forwarding frames to the appropriate output port(s). The forwarding decisions, however, are made by the IRC 40 located on the multiport switch 12.
According to the exemplary embodiment of the invention illustrated in FIG. 4, the IRC 40 includes four functional logic blocks, an ingress rules engine 200, a source address (SA) lookup engine 210, a destination address (DA) lookup engine 220 and an egress rules engine 230. In the exemplary embodiment, the four engines 200, 210, 220 and 230 are employed as separate logic devices. In other words, each engine is designed in a modular fashion to receive input from other devices and to perform its particular functions without relying on processing logic from another logic engine. Advantageously, this modular architecture allows changes to be made to any of the particular logic engines without affecting other parts of the decision making process. However, in alternative configurations, the individual functions performed by each logic engine, discussed in detail below, as well as the particular fnumber of logic engines may be modified, based on the particular network requirements. Additionally, in the exemplary embodiment, the entire IRC 40 is designed in a modular fashion, including a memory for storing frame headers and a scheduler for facilitating processing of the frame headers. Advantageously, storing the frame headers within the IRC 40 enables the IRC 40 to process the frames in an efficient manner without having to transmit handshaking signals to the respective MAC devices for forwarding the frame headers to the IRC 40.
The IRC 40 also includes address table 82. However, in alternative embodiments, the address table 82 may be located outside the IRC 40 within another part of the multiport switch 12 or even external to the multiport switch 12. According to the exemplary embodiment, the address table 82 supports 4096 user addresses and capabilities for 64 unique virtual local area networks (VLANs). However, the number of addresses and VLANs supported may be increased by expanding the table size. VLANs provide “broadcast domains” whereby broadcast traffic is kept “inside” the VLAN. For example, a specific VLAN may contain a group of users at a high level of an organization. When sending data to this group of users, the data may include a specific VLAN identifier associated with this particular group to ensure that only these users receive the data. These VLAN groupings can be thought of as “sub-networks” within a larger network.
FIG. 5 illustrates the organization of the IRC address table 82. The IRC address table 82 contains an array of 4096 entries. The first “n” entries 92 are referred to as “bin entries” and have addresses from “0” to “n−1”. The remaining entries 94 are referred to as “heap entries” and have addresses from “n” to “4095”. Each of the table entries includes a 72-bit address entry field and a 12-bit “next pointer” field.
FIG. 6 illustrates the composition of each 84-bit table entry shown in FIG. 5. The hit bit is used for address entry “aging” to delete entries from the address table 82 that have not been used in a predetermined amount of time. The static bit is used to prevent deletion of an address entry.
The traffic capture bit identifies traffic capture source and destination MAC addresses for mirroring MAC conversations to the management queue 68.
The VLAN index field is a 6-bit field used to reference a 12-bit VLAN identifier (ID). The VLAN index-to-VLAN ID table 86, shown in FIG. 4, contains the mapping associations. The switch 12 receives both tagged and untagged frames. When the switch 12 receives untagged data frames, i.e., without VLAN tag information, the IRC 40 assigns a VLAN index from the VLAN port-to-index table 88, shown in FIG. 4, based on the receive port on which the frame is received. The VLAN index-to-ID table 86 and the VLAN port-to-index table 88 are located with the configuration and status registers 52. However, in alternative configurations, the tables 86 and 88 may be located within the IRC 40.
The port vector is a 15-bit field that provides a forwarding descriptor with a vector identifying the port(s) to which the frame should be forwarded. The MAC address field is a 48-bit field that includes addresses for both source addresses and destination addresses. The addresses can be unicast, multicast or broadcast. An individual/group (I/G) bit is also included in the MAC address field.
In the exemplary embodiment of the present invention, the host CPU 32 functions as the management entity and is connected to the IRC 40 via the CPU IF 50. Alternatively, a management MAC may be connected to the CPU IF 50 to function as the management entity.
The IRC 40 uses the specific fields of the address table 82 to make frame forwarding decisions when frames are received in the switch 12. More specifically, the IRC 40 uses engines 200-230 to generate frame forwarding information and to create a forwarding descriptor for output to the port vector FIFO 56.
As discussed previously, the multiport switch 12 stores incoming data frames in external memory 36. According to the exemplary embodiment illustrated in FIG. 4, the IRC 40 also includes a logically separate 4-deep rules queue 120 allocated for each receive port, i.e., the queue corresponding to each receive port holds four frame headers. However, in alternative configurations, the rules queue 120 may be configured to store other numbers of frame headers for each port, based on the particular network requirements.
The rules queue 120 “snoops” on the write bus 69 to external memory 36 to capture a predetermined portion of the data frames, including the destination and source addresses, transferred by queuing logic 74 to the buffers in external memory 36. For example, the rules queue 120 may store the first 40 bytes of the frame. When a frame has been completely transferred to external memory 36, the queuing logic 74 signals the end of the transfer and provides frame status information indicating whether the frame was received at the switch 12 without errors. The IRC 40 also includes IRC scheduler 122, illustrated in FIG. 4, which monitors the signaling from queuing logic 74 and stores the frame status information in the rules queue 120 along with the corresponding frame header.
The rules queue 120 monitors the number of entries present at each port. When a queue for a receive port has three entries, the IRC 40 signals flow-control/back-pressure logic associated with that receive port in order to regulate network activity, the details of which are not disclosed herein in order not to unduly obscure the thrust of the present invention.
When the end of frame (EOF) transfer has been signaled by the queuing logic 74, the IRC scheduler 122 enables the processing of the frame header through the ingress rules engine 200. Logic engines 200-230, as discussed previously, are separate logic devices and are able to process data frames in parallel, thereby increasing data throughput as compared to systems which employ a single decision making device. In other words, each logic engine is able to perform its respective processing on a different data frame simultaneously with the other respective logic engines. Advantageously, the data throughput of the multiport switch 12 including engines 200-230 may increase up to fourfold, as compared to a network switch that employs a single decision making device, since four data frames may be processed simultaneously. The operation of each logic engine, according to the exemplary embodiment, will be described below.
The ingress rules engine 200 performs a variety of pre-processing functions for each frame header. For example, ingress rules engine 200 checks to see if a data frame was received with errors by reading the frame status information stored with the respective frame headers in rules queue 120. When the ingress rules engine 200 determines that a receive error has occurred, the ingress rules engine 200 constructs a forwarding descriptor with a null port vector, e.g., a port vector with all zeros or some other predetermined value, that will cause the frame to be discarded. Optionally, frames with errors may be forwarded to the host CPU 32 for diagnostic purposes.
The ingress rules engine 200 also checks the source address of the received frame to determine whether the Individual/Group (I/G) bit is set. If the I/G bit is set, indicating a multicast source address, the ingress rules engine 200 handles the frame as if the frame was received with errors. That is, the ingress rules engine 200 creates a forwarding descriptor with a null port vector.
The ingress rules engine 200 also checks the destination address (DA) of the frame to determine if the frame should be sent to the management entity, e.g., host CPU 32. Specifically, the ingress rules engine 200 looks for Bridge Protocol Data Units (BPDUs), Generic Attribute Registrations Protocol (GARP) frames, MAC Control Frames and frames with certain Physical MAC addresses. The ingress rules engine 200 identifies these types of frames based on their specific destination address information. When the ingress rules engine 200 detects a match with one of the above DAs, the ingress rules engine 200 constructs a forwarding descriptor identifying the management port as the forwarding port.
The ingress rules engine 200 also determines whether SA and DA lookups will be performed by engines 210 and 220, respectively, based on whether learning and forwarding are set in the respective port IRC control registers 114 a-m, illustrated in FIG. 4, in addition to whether the data frame was received with errors or is one of the specific types of frames to be transmitted to the management port, discussed above. According to the exemplary embodiment of the invention, the multiport switch 12 includes one port IRC control register 114 for each of the twelve 10/100 Mb/s ports and for the 1 Gb/s port. In alternative configurations, a single register could be used to store the appropriate control information for the respective ports.
Referring to FIG. 4, each port IRC control register 114 includes a learn bit and a forward (frwrd) bit. A set learn bit allows the IRC to “learn” unknown MAC source addresses received by the corresponding port, i.e., add new entries not stored in address table 82. A set frwrd bit allows frames received by the corresponding port to be forwarded to other ports and allows frames received by other ports to be transmitted from this port. When learning is set and forwarding is not set in the port IRC control register 114 corresponding to the port on which the frame was received, only the SA lookup is performed. That is, the SA lookup is performed so that a new entry may be added to the address table 82 and the SA lookup engine 210 generates a forwarding descriptor with a null port vector. When learning and forwarding are both set in the port IRC control register 114 corresponding to the receive port, both SA and DA lookups are performed, as discussed in more detail below. When learning and forwarding are both clear in the port IRC control register 114 corresponding to the receive port, neither the SA nor DA lookups is performed. In this case, the ingress rules engine 200 generates a forwarding descriptor with a null port vector.
Optionally, the ingress rules engine 200 performs VLAN ingress filtering to prevent the multiport switch 12 from forwarding a frame that does not belong to a VLAN associated with the receiving port. The port IRC control registers 114 each include an ingress bit which, when set, indicates that ingress filtering is enabled. Ingress filtering according to the exemplary embodiment of the present invention proceeds as follows.
Initially, the ingress rules engine 200 determines whether a received frame has no VLAN tag header or if the VLAN tag header has a VLAN ID equal to “0”. When the frame has no VLAN tag header or the VLAN ID is “0”, the ingress rules engine 200 does not perform ingress filtering regardless of the state of the ingress bit. Otherwise, the ingress rules engine 200 retrieves the VLAN index corresponding to the frame's VLAN ID from the VLAN index-to-ID table 86. If the frame's VLAN ID is not found in this table, the ingress rules engine 200 forwards the frame to the management port only.
After processing by ingress rules engine 200, the IRC 40 performs SA and DA searches of address table 82, based on whether learning and forwarding are enabled as discussed above. The multiport switch 12 needs to make frame forwarding decisions relatively quickly, since multiple data frames may be received by the multiport switch 12 simultaneously. Hence, in the exemplary embodiment of the present invention, a hashing scheme is used to search only a subset of the address entries, as described below. The memory structure of FIG. 5 provides an indexed arrangement, where a given network address will be assigned to a corresponding bin. In other words, each bin entry 96 is configured to reference a plurality of table entries (i.e., heap entries) 98. Hence, the SA lookup engine 210 performs a search of the address table 82 by first accessing a specific bin 96 pointed to by a hash key, and then searching the entries within (i.e., referenced by) the corresponding bin to locate the appropriate match.
Each bin entry 96 is the starting point for the search by the SA lookup engine 210 for a particular address within the address table 82. A bin entry may reference no addresses (i.e., be empty), may reference only one address within the bin entry location, or may reference a plurality of addresses using a linked list chain structure.
FIG. 7 is a diagram illustrating bin entries referencing a different number of table entries. Each of the bin entries 96 and heap entries 98 includes the 72-bit address entry and a 12-bit “next pointer” field. The “next pointer” field associated with the bin entry 96 identifies the location of the next entry in the chain of linked list addresses. For example, Bin 3, 96 d, of FIG. 7 does not have any associated table entries. In such a case, the 72-bit address entry equals zero (or another null value), and the bin's corresponding “next pointer” field will have a value of “1”, indicating no entries for the corresponding bin. If a bin such as Bin 1, 96 b, contains a single table entry, the bin entry will store the switching logic data for that single address in its address entry field, and store the value “zero” in the “next pointer” field, indicating there are no further address entries in the chain. Bin 0, 96 a, however, references four addresses by using the “next pointer” field to identify the location of the next entry in the chain. The additional entries 96 b-96 d in the bin are linked in a linear list, as shown in FIG. 7. Thus, the first entry of Bin 0 is stored in the address entry field of the bin entry 96 a and the next entry (heap entry 98 a) is referenced by address entry “a” in the next pointer field of the bin entry 96 a.
The SA lookup engine 210 performs hash searches of the IRC address table 82 to find entries associated with the source address and VLAN index of a received data frame. FIG. 8 is a block diagram illustrating an exemplary hash function circuit 100 used in conjunction with the SA lookup engine 210 in accordance with an embodiment of the present invention. The hash function circuit 100 includes a series of AND gates 102, a series of exclusive OR (XOR) gates 104, and a shift register 106. A user-specified hash function, stored in a user-programmable register (HASHPOLY) 108, includes a 12-bit value defining the hash polynomial used by the hash function circuit 100. Exemplary hash polynomials for the hashing function of the present invention are x12+x10+x7+x3+x2+1, which has a HASHPOLY of 0100 1000 1101, and x12+x10+x5+x3+1, which has a HASHPOLY of 0100 0010 1001. The x12 term is assumed to always equal “1”, and therefore is not stored in the HASHPOLY register 108. Other polynomials may also be used for HASHPOLY based on the particular design requirements.
The hash function circuit 100 generates the hash key using the source address of the data packet according to a user-specified hash function. Initially, the IRC controller 82 concatenates the 16 least significant bits of the source address of the data packet with the VLAN index to create a search key. After the entire search key has been processed, the hash function circuit 100 outputs a 12-bit has key.
From the 12-bit hash key, the SA lookup engine 210 calculates a bin number for searching the appropriate bin list in address table 82. More particularly, the SA lookup engine 210 uses the lower POLYEN bits of the hash key to generate the bin number. The bin number falls in the range of [0, n−1] where n=2POLYEN and the value of POLYEN is programmed by the host CPU 32 and stored in register 110. The hash key output by the hash function circuit 100 is provided to a logic circuit, for example a 12-bit parallel AND gate 111, that selectively outputs the lower significant bits of the hash key based upon a polynomial enable value (POLYEN) stored in register 210. The field “POLYEN” defines how many bits of the hash key are used to create the bin number. For example, if POLYEN=5, then the SA lookup engine 210 uses the lower five bits of the hash key. Hence, the hash key output by the logic circuit 100 is based upon masking the 12-bit hash key using the stored register value POLYEN in register 110.
After the bin number is calculated, the SA lookup engine 210 searches the bin list of the particular bin for an address entry whose address and VLAN index and fields match the source address (SA) and VLAN index of the received frame.
If the SA lookup engine 210 finds an address entry whose address and VLAN index match the SA and VLAN index of the frame, the SA lookup engine 210 sets the hit bit for that address entry. Optionally, the hit bit may be used for address entry aging. If the SA lookup engine 210 does not find a match and learning is enabled, the SA lookup engine 210 constructs a new entry in the IRC address table 82 using the information from the received frame.
After the SA lookup engine 210 completes the search and adds a new entry, if necessary, the DA lookup engine 220 performs a search of the address table 82, assuming that forwarding is set in the corresponding port IRC control register 114. Specifically, the DA lookup engine 220 searches the address table 82 for an address entry whose address and VLAN index match the destination address (DA) and VLAN index of the frame. The DA lookup engine 220 uses the 12-bit hash function circuit 100, illustrated in FIG. 8, to generate a 12-bit hash key for the DA/AVLAN index search. The DA lookup engine 220 uses the lower POLYEN bits of the hash key to calculate the bin number in the address table 82. The DA lookup engine 220 then searches the appropriate bin list for a DA/VLAN index match in the address table 82. If a match is found, the DA lookup engine 220 uses the port vector field of the address entry and passes the port vector field information to the egress rules engine 230. When the DA lookup engine 220 cannot find a DA/VLAN index match, the frame must be “flooded” to all members of the VLAN. In this case, the DA lookup engine 220 sets the port vector to indicate that all member ports are to transmit the frame.
After, the DA lookup engine 220 generates the port vector, the egress rules engine 230 receives the port vector information along with the receive port number and VLAN ID information. The egress rules engine 230 then creates a forwarding descriptor for the frame, as discussed in detail below.
FIG. 9 illustrates the composition of the forwarding descriptor according to an embodiment of the present invention. Referring to FIG. 9, the priority class field is a one-bit field that indicates the output priority queue in which the frame pointer should be placed, e.g., high priority or low priority.
The port vector field is a 15-bit field that identifies each port(s) that should receive the data frame for transmission to its destination address. Bit 0 of the port vector field corresponds to Port 0 (the management port), bits 1-12 correspond to MAC ports 1-12 respectively (the 10/100 Mb/s ports), bit 13 corresponds to the gigabit port 24 and bit 14 corresponds to the expansion port 30.
The untagged set field is a 13-bit field that indicates which ports should remove VLAN tag headers before transmitting frames. The untagged set is obtained from an untagged set table. The Rx port is a four-bit field that indicates the port from which the frame was received.
The VLAN ID field is a 12-bit field that includes the VLAN identifier associated with the frame. The opcode is an 11-bit field that contains instructions about how the frame should be modified before transmission and information that the host CPU 32 can use for processing frames from the management queue. The frame pointer is a 13-bit field that contains the location of the frame stored in external memory 36.
When VLAN ingress filtering is set, the egress rules engine 230 performs VLAN member set checking. The egress rules engine 230 performs this check by examining the bit that corresponds to the frame's VLAN index in the VLAN member set table entry that corresponds to the output port. If this bit is not set, the egress rules engine 230 masks that port from the port vector. The egress rules engine 230 also performs a trunk mapping function when an output port is part of a “trunk”, i.e., a predetermined set of ports that are used to transmit data to the same destination. When a port is part of a trunk, the egress rules engine 230 determines which port in the trunk on which to transmit the data frame.
After the egress rules engine 230 generates the forwarding descriptor, the egress rules engine 230 outputs the forwarding descriptor to the port vector FIFO 56 for queuing, as shown in FIG. 3. As discussed above, the IRC 40 processes data frames in a pipelined manner to increase data throughput. Additionally, the pipeline is asynchronous so that the SA and DA lookup engines 210 and 220 are able to search the address table 82 for the desired information. In other words, the SA and DA lookup engines 210 and 220 have a variable amount of time to search the address table 82 and are not limited to a fixed search time. Advantageously, this enables the SA and DA lookup engines 210 and 220 to search the address table 82 in an efficient manner to locate the desired information.
Described has been a system and method for generating a forwarding descriptor in a network interface device. An advantage of the invention is that logic engines 200-230 are designed as separate logic devices and are able to process data frames independently. For example, the egress rules engine 230 may process a first frame while the DA lookup engine 220 is processing a second frame, the SA lookup engine 210 is processing a third frame and the ingress rules engine is processing a fourth frame. Advantageously, the throughput of the switch increases data throughput as much as fourfold, as compared to a switch that processes frames one frame at a time. Another advantage of the invention is that the modular architecture of the IRC 40 enables changes to be made to one of the logic engines without affecting the other logic engines. For example, when changes to the ingress rules engine 200 are desired, the changes may be made without causing corresponding changes to any of the engines 210, 220 or 230. Accordingly, the desired changes may be made in an efficient manner while minimizing the complexity of the changes. A further advantage of the invention is that the entire IRC 40 is designed in a modular fashion including rules queue 120 for storing frame headers. The use of the rules queue 120 within the IRC 40 reduces flow control/back pressure problems associated with processing frames received by the MAC devices. The snooping function of rules queue 120 also reduces the need for additional handshaking signals between the receive MACs and the IRC 40 when storing frame headers in the rules queue 120. Accordingly, the modularity of the entire IRC 40 enables changes to be made to any part of the IRC 40 while minimizing the effects to other parts of the switch 12.
In this disclosure, there is shown and described only the preferred embodiments of the invention, but, as aforementioned, it is to be understood that the invention is capable of use in various other combinations and environments and is capable of changes or modifications within the scope of the inventive concept as expressed herein.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5515376||Jul 19, 1993||May 7, 1996||Alantec, Inc.||Communication apparatus and methods|
|US5694554 *||Aug 31, 1995||Dec 2, 1997||Hitachi, Ltd.||ATM interface and shaping method|
|US5838684 *||Feb 22, 1996||Nov 17, 1998||Fujitsu, Ltd.||Low latency, high clock frequency plesioasynchronous packet-based crossbar switching chip system and method|
|US6058112 *||Dec 18, 1997||May 2, 2000||Advanced Micro Devices, Inc.||Internal rules checker diagnostic mode|
|US6324164 *||Feb 4, 1998||Nov 27, 2001||International Business Machines Corporation||Asynchronous transfer mode (A.T.M.) protocol adapter for a high speed cell switching system|
|US6359879 *||May 6, 1998||Mar 19, 2002||Avici Systems||Composite trunking|
|US6389025 *||Jun 5, 2001||May 14, 2002||Hitachi, Ltd.||Distributed type switching system|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US6744776 *||Oct 18, 2000||Jun 1, 2004||Advanced Micro Devices, Inc.||Servicing priority traffic in multiport network switch|
|US6804234 *||Mar 16, 2001||Oct 12, 2004||Advanced Micro Devices, Inc.||External CPU assist when peforming a network address lookup|
|US6834056 *||Jun 26, 2001||Dec 21, 2004||Occam Networks||Virtual local area network protection switching|
|US6917623 *||Mar 26, 2001||Jul 12, 2005||Advanced Micro Devices, Inc.||System and method for transferring data in a network device|
|US6957272 *||Jan 18, 2001||Oct 18, 2005||Alcatel Internetworking (Pe), Inc.||Stackable lookup engines|
|US6977930 *||Feb 14, 2000||Dec 20, 2005||Cisco Technology, Inc.||Pipelined packet switching and queuing architecture|
|US6980552 *||Aug 15, 2002||Dec 27, 2005||Cisco Technology, Inc.||Pipelined packet switching and queuing architecture|
|US6999455 *||Jun 29, 2001||Feb 14, 2006||Broadcom Corporation||Hardware assist for address learning|
|US7003705 *||Jun 17, 2004||Feb 21, 2006||Extreme Networks, Inc.||Ethernet automatic protection switching|
|US7079533 *||May 2, 2001||Jul 18, 2006||Advanced Micro Devices, Inc.||Systems and methods for bypassing packet lookups|
|US7307988 *||Nov 22, 2000||Dec 11, 2007||Samsung Electronics Co., Ltd.||Address search apparatus and method in ethernet switch|
|US7379470 *||Apr 28, 2003||May 27, 2008||International Business Machines Coproration||Combined and data compressed FIFO based arbitration for a non-blocking switch|
|US7450595 *||May 1, 2001||Nov 11, 2008||At&T Corp.||Method and system for managing multiple networks over a set of ports|
|US7480242 *||Feb 28, 2003||Jan 20, 2009||Pluris, Inc.||Pass/drop apparatus and method for network switching node|
|US7643486||Oct 3, 2005||Jan 5, 2010||Cisco Technology, Inc.||Pipelined packet switching and queuing architecture|
|US7675930||Mar 9, 2010||International Business Machines Corporaiton||Chip circuit for combined and data compressed FIFO arbitration for a non-blocking switch|
|US7710954 *||Jun 7, 2006||May 4, 2010||Broadcom Corporation||Cascading of gigabit switches|
|US7796612||Dec 21, 2005||Sep 14, 2010||Broadcom Corporation||Gigabit switch with frame forwarding and address learning|
|US8018937||Sep 13, 2011||Cisco Technology, Inc.||Pipelined packet switching and queuing architecture|
|US8194685||Nov 10, 2008||Jun 5, 2012||At&T Intellectual Property Ii, L.P.||Method and system for managing multiple networks over a set of ports|
|US8665875||Sep 12, 2011||Mar 4, 2014||Oracle International Corporation||Pipelined packet switching and queuing architecture|
|US20010037396 *||Jan 18, 2001||Nov 1, 2001||Mathieu Tallegas||Stackable lookup engines|
|US20020037006 *||Jun 29, 2001||Mar 28, 2002||Broadcom Corporation||Hardware assist for address learning|
|US20020114336 *||Apr 17, 2001||Aug 22, 2002||Yu-Chun Chow||Gateway apparatus for performing communication between WAN and LAN|
|US20030137940 *||Feb 28, 2003||Jul 24, 2003||Schwartz Steven J.||Pass/drop apparatus and method for network switching node|
|US20050099945 *||Apr 28, 2003||May 12, 2005||International Business Machines Corporation||Combined and data compressed fifo based arbitration for a non-blocking switch|
|US20060039374 *||Oct 3, 2005||Feb 23, 2006||David Belz||Pipelined packet switching and queuing architecture|
|US20060050690 *||Oct 31, 2005||Mar 9, 2006||Epps Garry P||Pipelined packet switching and queuing architecture|
|US20080212577 *||Feb 19, 2008||Sep 4, 2008||International Business Machines Corporation||Chip circuit for combined and data compressed fifo arbitration for a non-blocking switch|
|US20080285437 *||May 18, 2007||Nov 20, 2008||Adc Dsl Systems, Inc.||Ethernet protection switching system|
|U.S. Classification||370/422, 370/428, 370/389|
|Cooperative Classification||H04L49/354, H04L49/103, H04L49/90, H04L49/3027|
|European Classification||H04L49/10E, H04L49/90|
|May 28, 1999||AS||Assignment|
Owner name: ADVANCED MICRO DEVICES, INC., CALIFORNIA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:MERCHANT, SHASHANK;EGBERT, CHANDAN;CHIANG, JOHN M.;REEL/FRAME:010006/0068;SIGNING DATES FROM 19990527 TO 19990528
|Jun 18, 2007||REMI||Maintenance fee reminder mailed|
|Dec 2, 2007||LAPS||Lapse for failure to pay maintenance fees|
|Jan 22, 2008||FP||Expired due to failure to pay maintenance fee|
Effective date: 20071202